Skip to main content

Production-ready Python SDK for launching, monetizing, and orchestrating AI Agents via ERC-8004

Project description

0rca Agent SDK

PyPI version License: MIT

0rca Agent SDK is a production-ready framework for building Sovereign, Monetizable, and Orchestrated AI Agents. It transforms standard AI agents (CrewAI, Agno, LangChain) into independent economic actors on the blockchain.

🚀 Key Features

  • Autonomous Monetization (x402 Protocol): Built-in support for HTTP 402 "Payment Required" flows. Agents handle their own pricing and payment verification.
  • Sovereign Identity: Every agent instance is a self-sovereign entity with its own EVM wallet and identity.
  • ERC-8004 Integration: Native support for trustless agent communication and escrowed payments.
  • Multi-Backend support: Pluggable architecture for your favorite AI frameworks:
  • Agent-to-Agent (A2A) Messaging: Standardized protocol for agents to find, chat, and collaborate with each other.
  • Production Persistence: SQLite-based logging and state management for every request and payment.

📦 Installation

pip install 0rca-agent-sdk

🛠 Quick Start

1. Create your Agent Handler

from orca_agent_sdk import AgentConfig, AgentServer

def handle_prompt(prompt: str) -> str:
    # Your core agent logic (LLM calls, tool usage, etc.)
    return f"Processed: {prompt}"

# Configure your sovereign agent
config = AgentConfig(
    agent_id="my-sovereign-agent",
    wallet_address="0xYourCreatorWallet...", # Where you receive payments
    price=10.0,                              # Price per request
    token_address="0xTokenAddress...",       # USDC, CRO, etc.
    chain_caip="eip155:25",                  # Cronos Mainnet
    ai_backend="crewai"                      # Your preferred backend
)

# Start the agent server
server = AgentServer(config, handle_prompt)
server.run(port=8000)

2. Built-in Contract Interaction

The SDK comes bundled with ABIs for the 0rca Network protocol contracts:

from orca_agent_sdk import load_abi

# Load a specific ABI
escrow_abi = load_abi("AgentEscrow")
identity_abi = load_abi("IdentityRegistry")

📜 Included Contracts (ERC-8004)

The SDK includes support and ABIs for the following production-level contracts:

  • TaskEscrow: Manage task-based budgets and secure payouts.
  • AgentEscrow: Trustless holding of payments until agent delivery.
  • IdentityRegistry: Decentralized registration of agent identities.
  • ReputationRegistry: On-chain trust and performance tracking.
  • ValidationRegistry: Verification of agent outputs.

💼 Task-Based Orchestration

The SDK now supports task-centric workflows where payments are handled via TaskEscrow.

1. Orchestrator Creates Task

The orchestrator (or user) creates a task on-chain with a budget.

2. Dispatch Task

Send a request to the agent including the taskId.

curl -X POST http://localhost:8000/agent \
  -H "X-TASK-ID: 0xYourTaskId..." \
  -d '{"prompt": "Do work", "taskId": "0xYourTaskId..."}'

3. Agent Execution & Spend

The agent tracks the task context and automatically calls spend(taskId, agentId, amount) on the contract upon completion or per-step.

🤖 A2A Protocol

Agents can communicate using the built-in A2A endpoints. The SDK handles message headers, timestamps, and routing between agent servers.

# Example A2A message
curl -X POST http://localhost:8000/a2a/send \
  -H "Content-Type: application/json" \
  -d '{
    "to": "another-agent-id",
    "action": "chat",
    "payload": {"prompt": "Hello!"}
  }'

🏗 Architecture

  • /agent: The main public entrance gated by x402.
  • /a2a: Message routing for inter-agent communication.
  • orca_agent_sdk/core: Heart of the SDK handling payments, identity, and persistence.
  • orca_agent_sdk/backends: Adapters for different AI frameworks.

📂 Examples

Check the examples/ directory for full implementations:

  • examples/mcp-server-agent: Full-featured agent with MCP tools and paywalls.
  • examples/client_example.py: Client script demonstrating the payment flow.

🧪 Development & Testing

Run the test suite to ensure everything is working:

python -m unittest discover tests

📄 License

MIT License. See LICENSE for details.


Built with 💙 by 0rca Network

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

orca_network_sdk-1.0.0.tar.gz (68.8 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

orca_network_sdk-1.0.0-py3-none-any.whl (77.1 kB view details)

Uploaded Python 3

File details

Details for the file orca_network_sdk-1.0.0.tar.gz.

File metadata

  • Download URL: orca_network_sdk-1.0.0.tar.gz
  • Upload date:
  • Size: 68.8 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.13.11

File hashes

Hashes for orca_network_sdk-1.0.0.tar.gz
Algorithm Hash digest
SHA256 f9a912c0af06da6078761c57084f376e7eda97ad20b4d126e97ed2ae02a72a70
MD5 1f6d7239b4a72aa8fa03a77f934a107c
BLAKE2b-256 562f475f1629af3889b63d52bccbcb35c703cd6092a4f42833cfac1cab15f912

See more details on using hashes here.

File details

Details for the file orca_network_sdk-1.0.0-py3-none-any.whl.

File metadata

File hashes

Hashes for orca_network_sdk-1.0.0-py3-none-any.whl
Algorithm Hash digest
SHA256 4d85bfec462a759c46ad52349603006c055118ac1061b411a015f7d22176ee51
MD5 c67be5514ef97156d8377329ff376da9
BLAKE2b-256 78406addb459189e8f86ef2db260c3934b2e243002afe1ee5060182371f6f100

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page